Analysis of the structure, organization and sequence of nucleic acid molecules is important in the prediction, diagnosis and treatment of human disease and in the study of gene discovery, expression and development. One laboratory tool used in the analysis of nucleic acid molecules is the microarray or high density array (HDA), which is a microarray containing a large number of targets per square centimeter of array surface. The microarray provides the framework for immobilization of nucleic acid molecules for analysis on a rapid, large-scale basis. Microarrays generally include a substrate having a large number of positionally distinct nucleic acid targets attached to a surface of the substrate for subsequent hybridization to a nucleic acid target. The key to efficiently immobilizing nucleic acid molecules is the surface chemistry and the surface morphology of the microarrays substrate.
Microarrays have led to advances in biochemistry, chemistry and engineering that have enabled the development of a new gene expression assay. This “chip-based” approach utilizes microscopic arrays of cDNAs printed on glass substrates as high-density hybridization targets. Fluorescent target mixtures derived from total cellular messenger RNA (mRNA) hybridize to cognate elements on the array, allowing accurate measurement of the expression of the corresponding genes. A fundamental requirement for gene expression analysis using microarrays is a sensitive and robust method for detecting the hybridized sample to the target DNA immobilized on the array. When DNA microarrays are used to measure the relative expression of MRNA between two samples (e.g. experimental and control), the targets representing the two samples are each labeled with a different fluorescent dye, mixed and hybridized with the microarray. The ratio of the two dyes, which reflects the level of differential gene expression, is obtained by analyzing the array at the two different wavelengths. Therefore, to a large extent, the microarray performance depends on the optimal and accurate detection of fluorescence emitted by the fluorophores conjugated to the target molecules.
An important element for successful microarray expression analysis is the quality of the substrate onto which hybridization targets are spotted. Poor quality slides result in low nucleic acid binding efficiency, poor spot morphology and fluorescent background that is often both relatively high and non-uniform.
The surface of a substrate used for microarrays also must contain a suitable functional group for attaching target biomolecules such as DNA to the substrate surface. Target biomolecules such as DNA will not attach to a naked glass substrate. There are two general functionalities on glass substrates for attaching DNA to the substrate surface. One is a surface including an aldehyde functionality, which is used to covalently attach amino-modified DNA onto the surface by reaction with free aldehyde groups using Schiff's base chemistry. Another different type of functionalization of a substrate surface involves non-covalent attachment. Amine and lysine coated slides are two examples of many coatings that provide for non-covalent attachment of biomolecules such as DNA to the surface of a substrate. Another example of a coating that provides for noncovalent attachment of biomolecules is a silane coating, such as an amino propyl silane. One shortcoming of substrates including an aldehyde functional group is that the substrates typically must be rinsed with a reducing agent to reduce free aldehydes on the surface of the slide and prevent attachment of target biomolecules to locations on the substrate surface that do not contain biomolecules.
The surfaces of both organic and inorganic substrates can be modified by the deposition of a polymeric monolayer coating or film to construct biomolecular assemblies. In addition, surface modification can also be used to promote adhesion and lubrication, modify the electrical and optical properties of the substrate surface, and create electroactive films suitable for various optical and electronic sensors and devices.
As noted above, compounds with amine functionality have found extensive application in the preparation of surfaces for nucleic acid hybridization. Due to their ability to bond to a substrate with a hydroxide and their ability to bond to nucleic acids with an amine, silane compounds are useful as surface coatings that will effectively immobilize nucleic acids. One example of a silane used for biological assay preparation is gamma amino propyl silane (GAPS), which may be deposited by a variety of methods, including CVD, spin coating, spray coating and dip coating.
Fluorescence detection sensitivity is severely compromised by background signals, which may originate from endogenous sample constituents/surface to which the target is immobilized or from nonspecific hybridization of probes to the target. Generally, the nonspecific signals referred to as background, but not the intrinsic auto-fluorescence, can be eliminated by a high stringency wash of arrays after hybridization. The intrinsic auto-fluorescence of the arrays obscures the sensitivity of gene expression analysis to a large extent by hindering the detectability of the low-level specific fluorescent signals.
Attempts to diminish or eliminate auto-fluorescence by selecting filters that reduce the transmission of emission relative to excitation wavelength or by selecting filters that absorb and emit at longer wavelengths are partially successful, but still have limitations. Although narrowing the fluorescence detection bandwidth increases the resolution, it also compromises the overall fluorescence intensity detected.
While the present invention should not be limited by a particular theory of operation, it is believed that fluorescence is caused by an aptly conjugated electronic system in an organic molecule. There are multiple potential sources of auto-fluorescence. Auto-fluorescence could be due to trace impurities of fluorescent molecules that typically contain single or conjugated pi bonding. In addition, during storage or printing, adsorption and oxidation of some biological or chemical contaminants, could result in the emission of fluorescence.
With so many possible causative agents for autofluorescence, it appears that there are at least three possible ways to circumvent this major problem. A first way is to make an array without any contaminants. However, this is very difficult to achieve. A second way to eliminate autofluorescence is to wash out contaminants after arraying. A drawback of this approach is that this method may take long time and may end up either losing DNA targets partially or incomplete washing off the contaminants. A third way involves the use of relatively simple chemical means. Applicants have discovered a relatively rapid, reproducible and easily applicable method of substantially reducing autofluorescence on slides including a functional group for non-covalent attachment to a biomolecule.
Previous studies with reducing agents to eliminate auto-fluorescence from paraffin embedded tissue sections were found to significantly decrease deceptive false positive fluorescent signals. Hydrides are known to reduce the conjugated system in organic molecules. Sodium borohydride and sodium cyanoborohydride are mild reagents and hydride donors, which are known to reduce double bonds in conjugated systems. They both work well to reduce background signals in tissue section for histochemistry and cytochemistry studies. See, e.g., Schnell, S A, Staines, S A and Wessendorf, M W. J., Reduction of lipofuschin like auto-fluorescence in fluorescently labeled tissue, Histochemistry Cytochemistry, 1999. 47 (6), 719-30; Clancy, B and Cauller, L. J. J., Reduction of background auto-fluorescence in brain sections following immersion in sodium borohydride, Neuroscience Methods, 83, 1998. 97-102; Beisker, W Dolberate, F. and Gray J W, An improved immunocytochemical procedure for high sensitivity detection of incorporated bromodeoxyuridine, Cytometry 1987: 8: 235-9; and Willingham M C. J. Histochemistry Cytochemistry, Alternative fixation processing method for preembedding ultrastructural immunocytochemistry of cytoplasmic antigens, 1983: 31-791-889.
Taking into account the wide variety of factors contributing to autofluorescence, techniques for the elimination of autofluorescence in one biological system containing a specific type of organic molecules and chemicals would not be expected work in a system employing different organic molecules and chemicals. Furthermore, methods used in the elimination of autofluorescence problems in bulk tissue samples would not be expected to be useful in the elimination of autofluorescense on substrates used for microarrays. Other researchers have suggested eliminating autofluorescense on slides containing non-covalent attachment functionality through curing by baking at high temperatures. See, e.g., Super Microarray Substrates Handbook, Telechem International, Inc.//arrayit.com, www.arrayit.com, 1999. However, applicants experiments have found that curing of the slides in an oven was largely ineffective in reducing autofluorescence.
It would be useful to provide an improved method of treating substrates for immobilization and hybridization of biomolecules such as nucleic acids and oligonucleotides. The method should have the ability to be performed in a reproducible manner. It would also be advantageous to provide a substrate that has uniform surface characteristics and exhibits low background noise or autofluorescence when the substrate is analyzed using fluorescence scanning.